SLIM-LLMs: Modeling of Style-Sensory Language RelationshipsThrough Low-Dimensional Representations
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2025
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866912519727087616 |
|---|---|
| author | Khalid, Osama Srivastava, Sanvesh Srinivasan, Padmini |
| author_facet | Khalid, Osama Srivastava, Sanvesh Srinivasan, Padmini |
| contents | Sensorial language -- the language connected to our senses including vision, sound, touch, taste, smell, and interoception, plays a fundamental role in how we communicate experiences and perceptions. We explore the relationship between sensorial language and traditional stylistic features, like those measured by LIWC, using a novel Reduced-Rank Ridge Regression (R4) approach. We demonstrate that low-dimensional latent representations of LIWC features r = 24 effectively capture stylistic information for sensorial language prediction compared to the full feature set (r = 74). We introduce Stylometrically Lean Interpretable Models (SLIM-LLMs), which model non-linear relationships between these style dimensions. Evaluated across five genres, SLIM-LLMs with low-rank LIWC features match the performance of full-scale language models while reducing parameters by up to 80%. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2508_02901 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | SLIM-LLMs: Modeling of Style-Sensory Language RelationshipsThrough Low-Dimensional Representations Khalid, Osama Srivastava, Sanvesh Srinivasan, Padmini Computation and Language Sensorial language -- the language connected to our senses including vision, sound, touch, taste, smell, and interoception, plays a fundamental role in how we communicate experiences and perceptions. We explore the relationship between sensorial language and traditional stylistic features, like those measured by LIWC, using a novel Reduced-Rank Ridge Regression (R4) approach. We demonstrate that low-dimensional latent representations of LIWC features r = 24 effectively capture stylistic information for sensorial language prediction compared to the full feature set (r = 74). We introduce Stylometrically Lean Interpretable Models (SLIM-LLMs), which model non-linear relationships between these style dimensions. Evaluated across five genres, SLIM-LLMs with low-rank LIWC features match the performance of full-scale language models while reducing parameters by up to 80%. |
| title | SLIM-LLMs: Modeling of Style-Sensory Language RelationshipsThrough Low-Dimensional Representations |
| topic | Computation and Language |
| url | https://arxiv.org/abs/2508.02901 |